The EC Discussion Paper on the Application of Article 82 Pinar Akman LL.B., LL.M., Ph.D. Candidate, Norwich Law School, Centre for Competition Policy

Transcription

1 The EC Discussion Paper on the Application of Article 82 Pinar Akman LL.B., LL.M., Ph.D. Candidate, Norwich Law School, Centre for Competition Policy Introduction European Commission DG Competition has published the long-awaited discussion paper on the application of Article 82EC in late December, 2005 (the paper ). 1 It is part of the framework for reforming EC competition rules and aimed at provoking discussion before the adoption of Commission guidelines on modernising the practice of Article 82EC. The paper sets out the Commission s application of the provision to exclusionary abuses and leaves exploitative abuses for future work. The policy of the Commission and the jurisprudence on Article 82EC have been criticised for not being grounded on sound economics and the current formbased approach being inferior to an effects-based approach. The most comprehensive study arguing for an effects-based approach in this manner is the report by the Economic Advisory Group for Competition Policy commissioned by the DG Competition (the EAGCP Report ). 2 The paper, nevertheless, seems to have missed the opportunity of answering the criticisms in full and in some instances appears to be more of an attempt at legitimising the current approach. In general, it lacks the vigour found in the EAGCP Report. The aim of this note is thus to briefly analyse the paper, assess how much the criticisms towards the application of Article 82EC have been resolved and compare the proposed application with the status quo. An Effects-based Approach and Consumer Welfare An effects-based approach focuses on the presence of anti-competitive effects that harm consumers and is based on the examination of each specific case, based on p.akman@uea.ac.uk. The support of the Economic and Social Research Council (ESRC) and the Overseas Research Students Awards Scheme (ORSAS) is gratefully acknowledged. The author would also like to thank Dr. Michael Harker and Prof. Morten Hviid for helpful comments. The usual disclaimer applies. 1 DG Competition discussion paper on the application of Article 82 of the Treaty to exclusionary abuses Brussels, December 2005 can be found at 2 Report by the EAGCP on An Economic Approach to Article 82 can be found at See also Competition Law Forum Article 82 Review Group The Reform of Article 82: Recommendations on Key Policy Subjects (2005) 1 (1) European Competition Journal

2 sound economics and grounded on facts. 3 In such an approach, the main concern is the improvement of consumer welfare as discussed at great length in the EAGCP Report and this is suggested to result in the protection of competition rather than competitors. 4 Although consumer welfare is presented as the ultimate concern in the paper as well 5, nowhere in it is explained in detail how an assessment of the effects on consumers is to be made. Nevertheless, lip service is paid to adopting an approach which is based on the likely effects on the market. 6 Whereas the EAGCP Report emphasises the need for the competition authority to explain in each case the harm for consumers from the company s conduct and indicate precisely the relevant consumer welfare effects in its decision 7, the stance taken in the paper appears to be an implicit assumption that exclusionary behaviour necessarily harms consumers and the contrary can be proved as a defence by the company. 8 It is indeed presumed that harm to intermediate buyers generally means harm to final consumers. 9 If this is the correct interpretation of the paper, however, then it would mean that the Commission has still not made it clear enough how the protection of competition is to be differentiated from that of competitors. The statement in the introduction of the paper that [a] wide variety of practices may be abusive if carried out by an undertaking in a dominant position 10 shows that the focus may still be mainly on the type of practice, i.e. the form of conduct rather than the effects. Moreover, the opportunity to explain how trade-offs between the interests of different consumer groups will be made, if necessary, also appears to have been missed despite this issue being pointed out by the EAGCP. 11 In the same vein, the paper does not touch upon how the Commission would balance the false positives and false negatives in its policy since the welfare effects of some infringements of Article 3 EAGCP Report n. 2 above, p. 2. A form-based approach concentrates on the type of conduct and is problematic since alternative practices can serve the same purpose, ibid p EAGCP Report n. 2 above, pp. 2, 7 et seq. 5 See e.g. paras. 4, 54, 55, Para EAGCP Report n. 2 above, pp. 8-9, See e.g. para Para. 55. It is not only short term harm, but also medium and long term harm to be taken into account, ibid. 10 Para EAGCP Report n. 2 above, p

3 82EC are ambiguous. 12 In other words, there is no mention of the costs of possible over- and under-intervention by the Commission. The recognition in the EAGCP Report that competition itself is the best mechanism for avoiding inefficiencies and the authority should not try to let its own intervention replace the role of competition 13 is thus not found in the paper. Relationship with other Treaty Provisions The paper starts out with the relationship between Article 82EC and other Treaty provisions. The application of Article 82EC in conjunction with Article 86EC and the specific situations likely to arise out of Article 10EC are left outside the scope of the paper. The relationship between Article 82EC and Article 81EC is nevertheless shortly explained. Although it mostly draws on existing Commission Guidelines 14, the conclusion reached in the paper blurs the issue rather than clarifying it. The paper states that if the conduct of a dominant company generates efficiencies and provided that all the other conditions of Article 81 (3) are satisfied such conduct should not be classified as an abuse under Article 82 of the EC Treaty. 15 This is a confusing assertion in that it is not clear whether what is meant by it is simply that restrictive agreements entered into by dominant firms can nevertheless be exempted under Article 81(3) (provided that other conditions of that provision are met) 16 or whether abusive behaviour by a dominant firm can be exempted under Article 81(3). A literal reading of the paper would suggest the latter understanding and hence mean the introduction of a rule of reason approach to Article 82EC. This would, however, be too radical a change to bring on within half a paragraph. Nevertheless, the submission in the EAGCP Report that an effects-based analysis would naturally lend itself to a rule of reason approach to competition policy suggests that this may well be what the Commission intends. The proposed application of the efficiency defence, 12 EAGCP Report argues that the likelihood of false positives (condemning a pro-competitive practice in a particular case) and false negatives (allowing a dominant company to abuse its position in other cases) and their costs should be balanced by the authority. Since both types of errors are likely, they would be better assessed by a case-by-case approach (n. 2 above, p. 7). 13 EAGCP Report n. 2 above, p Commission Notice, Guidelines on the Application of Article 81 (3) of the Treaty [2004] OJ C101/ Para This is the current position adopted on the Guidelines on the Application of Article 81 (3) n. 14 above, para The discussion paper itself asserts that consistency requires that Article 81 (3) EC be interpreted as precluding any application of this provision to restrictive agreements that constitute an abuse of a dominant position, para. 8. 3

4 as argued below, 17 overall the paper also seems to favour such an interpretation. Thus, a clearer and more thorough explanation would be welcome in the following Commission guidelines on the subject. Market Definition It is recognised in the paper that the concept of dominance relates to a position of economic strength on a market and thus, the application of Article 82EC requires the definition of a relevant market whose main purpose is to identify in a systematic way the immediate competitive constraints faced by the undertaking. 18 The paper generally refers to the existing Commission Notice in this regard. 19 It nevertheless provides additional guidance in one specific issue, i.e. the so-called cellophane fallacy 20. Although the Commission s current practice has been criticised for drawing markets too narrowly, the paper ironically - merely discusses the cellophane fallacy which works as a warning for the authority not to draw markets too widely. Hence, the criticisms do not seem to have been regarded in this matter. The paper does, however, set out that no single method of market definition, including the SSNIP test is likely to be adequate. A variety of methods have to be relied on for checking the robustness of possible alternative market definitions. 21 An alternative approach is suggested which requires the examination of the characteristics and intended use of the products concerned and to assess whether the consumer demand is inelastic. It is made clear that regard should be had to the needs of marginal consumers during this assessment and what matters is that a sufficiently large number of consumers consider one product to be a good substitute for the dominant 17 See text to n. 73 below. 18 Paras Commission Notice on the definition of the relevant market for the purposes of Community competition law [1997] OJ C372/3. 20 Cellophane fallacy refers to the scenario where the price charged by the allegedly dominant undertaking will have been raised above the competitive market level and hence the SSNIP test will fail to be of guidance in defining the market. The SSNIP test asks whether the customers of the allegedly dominant undertaking would switch to readily available substitutes or to suppliers elsewhere to such an extent that it would be unprofitable to implement a small but significant non-transitory increase in prices for the products. As long as the answer is in the affirmative, new products are added to the market definition until a market is found where it would be profitable to increase the price in this manner. See paras Para. 13. The appropriate benchmark for the assessment in the SSNIP test is given as the competitive price which is not necessarily the prevailing price in case of dominance, para. 15. It is further stated that although a method of avoiding the cellophane fallacy could be reconstructing the competitive price, in most cases it would not be possible to reconstruct the competitive price with sufficient accuracy, para

5 undertaking s product. 22 This is welcome in that the Commission recognises the possible insufficiency of the SSNIP test and proposes an alternative approach. 23 The paper also asserts that a comparison of prices across regions may help in the determination of both the geographic and the product market. It is stated that where an undertaking is able to set different prices in different regions depending on its share of sales (higher prices corresponding to higher share) of a certain product, this would be an indication that the main competitive constraint comes from other suppliers of that product rather than suppliers of different products. Moreover, even if the undertaking in question does not itself supply its product in different regions, a similar analysis can still be made if the same type of product is sold in other regions by other undertakings. 24 In other words, the fact that the undertaking is able to adjust its price according to its respective share of sales in two different regional markets implies that all else equal - the pricing of that product is independent of the prices of other products and thus, constitutes a market of its own. Such an assessment heavily relies on the similarity of the competitive conditions in the different regions which is a factor considered in the paper as well. If there are region specific reasons allowing the undertaking to charge different prices, it should not be easily concluded that other products do not constitute a competitive constraint. Dominance The paper appears not to have accepted the approach proposed in the EAGCP Report on the issue of dominance, but to have preferred to stay loyal to the status quo. The EAGCP Report s position is that an effects-based approach does not require a separate assessment of dominance since when a consistent and significant competitive harm is identified using that approach, that in itself would be evidence of dominance. 25 Accordingly, when an effects-based approach provides evidence of an abuse which is only possible if the company is dominant, no further demonstration of 22 Para. 18. In this case, two products would be considered to be in the same market unless a single supplier is able to price discriminate between the groups according to their demand elasticity and prevent arbitrage from the more elastic to the less elastic demand group. In the latter instance, the group with less elastic demand may constitute a market of its own. See also Notice on market definition, n. 19 above, para See R. Whish, Competition Law (5 th ed., Butterworths, London, 2003), p. 31 and A. Jones and B. Sufrin EC Competition Law (2 nd ed., Oxford University Press, Oxford, 2004), p. 58 criticising the Commission notice on the definition of the relevant market for not explaining what method would be used when the SSNIP test is not helpful due to the cellophane fallacy. 24 Para EAGCP Report n. 2 above, pp. 4, 14. 5

6 dominance should be needed. 26 The Commission seems to have followed rather the caveat that such a presumption may not hold in all cases 27 and thus preserved the interpretation more consistent with the jurisprudence. In defining dominance, the paper adopts the current definition of settled case law and restates that it is a position of economic strength enjoyed by an undertaking which enables it to prevent effective competition being maintained on the relevant market by affording it the power to behave to an appreciable extent independently of its competitors, its customers and ultimately of the consumers. 28 It is stated that an undertaking must have substantial market power to be dominant and market power in this sense means the power to influence market prices, output, innovation, the variety or quality of goods and services, or other parameters of competition on the market for a significant period of time. 29 After defining the market, whilst analysing dominance, it is relevant to consider especially the market position of the allegedly dominant undertaking, the market position of competitors, barriers to expansion and entry, and the market position of buyers. 30 The paper then continues with separate assessments of these factors. Market share is recognised as only a proxy for market power which is the decisive factor and attention is drawn to the importance of product differentiation, in that the competitive constraint that differentiated products impose on each other is likely to differ even if they are parts of the same relevant market. 31 Nevertheless, the current state is confirmed; it is very likely that very high market shares held for sometime are indicative of dominance. Similarly, the market share range of 40 % to 50 % is still the critical range as regards a finding of dominance. 32 At a market share of 50 % dominance is almost assumed provided that rivals hold a much smaller share of the market. Thus, the criticism of the EAGCP Report that structural indicators traditionally used as proxies for dominance do not always provide an appropriate measure of power 33 is not fully reflected in the paper. 26 EAGCP Report n. 2 above, p See J. Vickers Abuse of Market Power (2005) 115 The Economic Journal F244, F246; Sufrin and Jones n. 23 above, p Para. 20. See Case 27/76 United Brands Co and United Brands Continental BV v Commission [1978] ECR 207, para. 65; case 85/76 Hoffmann-La Roche & Co AG v Commission [1979] ECR 461, para Paras Para Paras Para EAGCP Report n. 2 above, pp

7 As for barriers to expansion and entry, it is rightly recognised that if the barriers faced by competitors and potential competitors are low, a high market share may not be indicative of dominance. 34 The Commission will look at whether any expansion or entry has been or would have been or will be sufficiently immediate and persistent to prevent the exercise of substantial market power. 35 These barriers are defined as factors that make entry impossible or unprofitable while permitting established undertakings to charge prices above the competitive level. 36 It is worth noting here that unlike the currently used concept it is not only barriers to entry, but also the barriers to expansion of current competitors that will be taken regard of. Although the given definition is appropriate for barriers to entry, the same cannot be said for barrier to expansion. It can be inferred from the paper though, barriers to expansion would be factors which make expansion of existing competitors impossible or unprofitable. It must be borne in mind, however, that such factors may be due to, for example, the inefficiency of existing competitors. Therefore, it would perhaps be more helpful to assess whether the expansion of a competitor at least as efficient as the dominant company is unprofitable or impossible. 37 According to the paper, expansion or entry which is not of sufficient scope and magnitude is not likely to be an effective constraint on the undertaking concerned. 38 The paper thus differentiates between different sizes of expansion or entry. Therefore, it is not enough that entry is possible or existent and the threat of entry is not spelled out as a constraint in itself on the dominant company. Such a proposition is not in total conformity with the way the paper defines barriers to expansion and entry since any type of entry which has actually happened and is profitable would satisfy the criteria in the definition. 39 Moreover, as the paper itself admits [a]ny attempt by an undertaking to increase prices above the competitive level would attract expansion 34 Para Para. 35. After stating that the period of time needed to decide whether expansion or entry is immediate depends on the characteristics and dynamics of the market, it is argued that the time undertakings already active on the market would need to adjust their capacity can be used as a yardstick to determine this period, para Para. 38. The paper names possible origins of barriers to expansion or entry as legal barriers, capacity constraints, economies of scale and scope, absolute cost advantages, privileged access to supply, a highly developed distribution and sales network, the established position of the incumbent firms on the market and other strategic barriers, para See Vickers n. 27 above, F256 et. seq. for an explanation of the as-efficient competitor test. See also text to n. 60 below. 38 Para See text to n. 36 above. 7

8 or new entry by rivals thereby undermining the price increase. 40 By bringing a quantitative and qualitative assessment of entry, the paper introduces a somewhat arbitrary test since it will be necessary to determine what constitutes an entry of sufficient scope and magnitude and what does not. Furthermore, this should be dealt with carefully since even though an entry on its own may not seem to be of sufficient scope and magnitude, if there is still the possibility of limitless entry, the total effect of this multitude of entries as a threat to the incumbent would still be significant. Analysis of Exclusionary Abuses The essential objective of Article 82EC with regard to exclusionary abuses is cited as that of protecting competition on the market as a means of enhancing consumer welfare and of ensuring an efficient allocation of resources. 41 It is clearly stated that the concern is to prevent exclusionary conduct of the dominant undertaking which has the potential to limit the remaining competitive constraints so as to avoid consumer harm. Hence, it is competition rather than competitors to be protected. 42 The central consideration is thus foreclosure that hinders competition and thereby harms consumers. 43 Exclusionary abuses are defined as behaviours by dominant firms which are likely to have a foreclosure effect on the market, i.e. which are likely to completely or partially deny profitable expansion in or access to a market to actual or potential competitors and which ultimately harm consumers. 44 This understanding certainly hinges on how one interprets likely and the broader that interpretation, the more likely that conduct will be found abusive. Moreover, it is not clear at first sight whether harm to consumers is part of the definition of foreclosure or whether it is a condition in addition to foreclosure to deem a behaviour exclusionary. From the later definition of foreclosure in the paper as the complete or partial denial of actual or potential competitors profitable access to the market 45, it is understood that harm to consumers is an additional condition. Hence, harm to consumers is an additional factor to foreclosure to find behaviour exclusionary which would mean that the 40 Para Para Para Para Para Para. 58. The paper states that foreclosure may discourage expansion/entry or encourage exit. It is sufficient that the competitors are disadvantaged and thus led to compete less aggressively for a finding of foreclosure; hence, it is not necessary that they are forced to exit, ibid. 8

9 differentiation between the prevailing categories of abuse, i.e. exclusionary and exploitative, is no longer sustainable since exclusionary would as a result of the Commission s understanding, require exploitation of consumers as well. This could thus be interpreted as a move towards a single type of abuse, i.e. exploitative. 46 Indeed, the expression of the essential objective of Article 82EC the protection of competition on the market as a means of enhancing consumer welfare and of ensuring an efficient allocation of resources surely supports this interpretation. 47 This is mostly significant when a conduct is possibly of an exclusionary nature, but does not necessarily have exploitative effects. If such conduct is not to be sanctioned under Article 82EC, then this would mean that there is only one type of abuse, i.e. exploitative. Nonetheless, the paper later on argues for the conformity of its definition with that of the ECJ in Hoffmann-La Roche which does not look for harm to consumers. 48 Moreover, as that conception of abuse rests on the influence of the behaviour on the market structure, the paper explains that the foreclosure capability of a conduct can in general be established by investigating the form and nature of the conduct. 49 Although it continues to look for an establishment of likely market distorting foreclosure effect, it raises concern with regard to how much emphasis is to be put on the form as opposed to the effect of the conduct. Thus, the following guidelines would be welcome to clarify this link between exclusionary and exploitative abuses. The Test for Exclusionary Abuses The test for the prohibition of exclusionary abuses in Article 82EC is suggested to be one that looks for both actual or likely anti-competitive effects in the market and direct or indirect consumer harm. 50 Along with the presumption that harm to 46 See E. M. Fox What is Harm to Competition? Exclusionary Practices and Anticompetitive Effect (2002) 70 Antitrust Law Journal 372 questioning whether there is ultimately only one type of abuse, that is exploitative. 47 Para. 4. The same understanding is expressed to be the aim of EC competition rules in general in Commission Guidelines on the application of Article 81 (3) n. 14 above, para Abuse is defined there as as an objective concept relating to the behaviour of an undertaking in a dominant position which is such as to influence the structure of a market where, as a result of the very presence of the undertaking in question, the degree of competition is weakened and which, through recourse to methods different from those which condition normal competition in products or services on the basis of the transactions of commercial operators, has the effect of hindering the maintenance of the degree of competition still existing in the market or the growth of that competition Hoffmann-La Roche, n. 28 above, para Para Para

10 intermediate buyers would generally create harm to final consumers 51 and the abovementioned 52 tendency towards a single type of abuse, there are two implications of this test. First, it is not for the sake of competition, let alone the sake of competitors, that competition is protected; even with regard to exclusionary conduct, harm to consumers appears to be necessary. Secondly, if harm to intermediate buyers is generally presumed to harm consumers, in private enforcement of competition law, this would mean that consumers would be able to sue the dominant undertaking for its anti-competitive behaviour on the intermediate market. In other words, consumers could have standing although they are not directly harmed by the conduct distorting the competitive conditions on the intermediary market. This would broaden the scope of private enforcement of EC competition law to a greater extent than that in the US since direct antitrust injury is needed in the latter. 53 The test also signifies a crucial difference between the EAGCP Report and the paper. In the EAGCP Report, the test of competitive harm is a consumer welfare test and competition is deemed to be harmed only when consumers are harmed. 54 Hence, the EAGCP Report takes the position that the standard for assessing whether a conduct is detrimental to competition or whether it is legitimate competition should be derived from the effects of the practice on consumers and thus, it is incumbent on the authority to explain in each case the harm to consumers. 55 It is argued there that otherwise, the authority may be tempted to confuse the protection of competition with the preservation of a particular market structure which would merely have the effect of protecting the competitors. 56 Contrarily, the paper looks for anti-competitive effects in the market and possible harm to consumers. In other words, they are not seen as one and the same thing. Moreover, although the avoidance of harm to consumers is seen as the ultimate concern of the provision, the test proposed in the paper does not look for proof of any actual or possible harm to consumers. 57 Such an approach brings 51 Para See text to n. 46 above. 53 For the position in the US see Illinois Brick Co. v. Illinois, 431 U.S. 720 (1977), Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc. 429 U.S. 477, 489 (1977) and Blue Shield of Virginia v. McCready, 457 U.S. 465 (1982). Cf EC Commission Green Paper Damages actions for breach of the EC antitrust rules COM(2005) 672 final. 54 EAGCP Report n. 2 above, pp EAGCP Report n. 2 above, pp EAGCP Report n. 2 above, p As stated at text to n. 9 above, the paper does include a presumption that harm to intermediate buyers generally means harm to final consumers. Although this is stated under the heading The Central Concern and Proof of Foreclosure, intermediate buyers are usually not the concern for foreclosure 10

11 with it the danger of interpreting harm to competitors as harm to consumers and thus, one could end up protecting competitors rather than consumers. The paper unfortunately fails to sufficiently elaborate on this argument. Another important position of the paper is the one that [w]here a certain exclusionary conduct is clearly not competition on the merits, in particular conduct which clearly creates no efficiencies and which only raises obstacles to residual competition, such conduct is presumed to be an abuse. 58 Nevertheless, the dominant undertaking is given the possibility to rebut this presumption by providing convincing evidence that the conduct does not and will not have the alleged likely exclusionary effect or that the conduct is objectively justified. This approach raises concern in that a presumption of abuse is suggested which may be dependent on the form of conduct, rather than its effect. This is because unlike the wording of the paper, it is not necessarily clear what constitutes competition on the merits and what does not. Competition on the merits is somehow a vague concept 59 and a presumption of abuse based on it may result in the finding of some particular conduct as abusive and hence, reduce the test to a form-based one; whenever the authority is faced with that particular conduct, it may assume that it is abusive. The paper differentiates between price-based and non-price based exclusionary behaviour and asserts that for the former, the principles set out in the paper are based on the premise that in general only conduct which would exclude a hypothetical as efficient competitor is abusive. 60 The hypothetical as efficient competitor is one with the same costs as the dominant company. It is moreover stated that foreclosure of an as efficient competitor can in general only result if the dominant company prices below its own costs. The question to be asked for price-based exclusionary abuses is thus, whether the dominant company itself would be able to survive the exclusionary conduct in the event that it would be the target. 61 This is the effects since they are generally not in competition with the upstream dominant company. As recognised in the paper itself, from a competition policy point of view, this is mostly only a worry if the dominant company is itself active downstream (para. 72). 58 Para See Jones and Sufrin n. 23 above, pp for the argument that the enduring difficulty with Article 82EC is where the courts and the Commission draw the line between anti-competitive conduct and competition on the merits. See also Whish n. 23 above, p. 197 for the same argument. 60 Para Para. 66. The Commission makes clear that it is necessary to look at the revenues and costs of the dominant company in a wider context to see whether the dominant company s conduct negatively affects its revenues in other markets or of products other than the one under scrutiny, para 67. Moreover, in case reliable information on the dominant company s costs is not available, cost data of 11

12 current approach already followed, for example, in margin squeeze cases. 62 The hypothetical as efficient competitor test may nevertheless be misleading since the dominant company may not be efficient exactly because it is abusing its dominance, i.e. because it is pursuing a quiet monopoly life. It may be better to ask, therefore, whether conduct would exclude an efficient competitor or one that is at least as efficient as the dominant company. In the context of the as efficient competitor test, the paper posits that it may sometimes be necessary in the consumers interest to also protect competitors that are not (yet) as efficient as the dominant company. 63 Thus, it recognises that the protection of competitors may at some times be intertwined with the protection of competition. The EAGCP Report also notes that in some cases concerns for the protection of competitors from certain forms of inappropriate behaviour may be appropriate. The caveat is that this is not true for all cases and competitors themselves should not be protected from competition by intervention; the authority must assess these matters without any prejudice for any particular structure. 64 Objective Justification The paper sheds light to the issue of objective justification by stating that exclusionary conduct may escape the prohibition of Article 82EC if the dominant undertaking can provide an objective justification for its behaviour or it can demonstrate that its conduct produces efficiencies which outweigh the negative effect on competition. 65 Hence, the dominant company can either prove an objective justification or an efficiency defence to defend its conduct. Two types of objective justification are possible. First, the objective necessity defence under which the dominant undertaking is allowed to show that the otherwise abusive conduct is actually necessary on the basis of factors external to the parties involved, and especially to the dominant undertaking. The conduct must be generally necessary for all undertakings and so indispensable that without it the products concerned cannot or will not be produced or distributed in that market. 66 Secondly, apparently efficient competitors may be used instead. In case no reliable information on cost data is available, but the Commission has nevertheless been able to build on other arguments a credible case of abuse, the dominant company may show that it is not pricing below the appropriate benchmark, ibid. 62 See Commission Decision Deutsche Telekom AG (2003/707/EC) [2003] OJ L263/9, paras Para. 67. See Vickers n. 27 above, F256 for the argument that the entry of less-efficient competitors can improve social welfare in certain circumstances. 64 EAGCP Report n. 2 above, p Para Para

13 the meeting competition defence where the dominant undertaking is able to show that the otherwise abusive conduct is actually loss minimising reaction to competition from others. 67 The paper sets out that the meeting competition defence can only be used for behaviour which would otherwise be a pricing abuse and it can only apply to individual behaviour to meet competition. 68 Behaviour should moreover be proportionate to make use of a meeting competition defence. 69 The Commission s understanding of defending one s own commercial and economic interests in the face of competitors actions, as that is the concept used by the Community courts, appears to be that of minimising short run losses resulting directly from competitors actions. 70 In other words, the dominant company is only allowed to react in response to a lowering of prices by competitors. Beating competition, rather than meeting it still seems to be not allowed for a dominant firm. To pass the proportionality test, the dominant company must be able to show that the conduct is a suitable way to achieve the legitimate aim, is indispensable and that meeting competition is a proportionate response in view of the aim of Article 82EC. 71 It is stated that with a view to protect the consumers interests, the proportionality test requires a case by case weighing of the interest of the dominant company to minimise its losses and the interest of its competitors to enter or expand. This last statement is not easy to comprehend since it is difficult to see the link between preventing a dominant company from reacting to price cuts by competitors at the expense of making losses and the protection of consumers. If the dominant company would not be allowed to lower its prices to meet the competition from competitors, this would result in some prices being higher than what would otherwise be and interfere with the competitive process. Moreover, it would amount to onerously asking the dominant firm to lose business. Furthermore, it again seems to assume that interests of competitors and their protection are necessarily those of consumers as well. Nonetheless, as mentioned above, 72 the protection of the interests of competitors would only in certain cases be appropriate and it is only limited to the 67 Para Para Para. 81. See Jones and Sufrin n. 23 above, p. 283 for the argument that the application of proportionality to the facts of any given case does not help to predict the outcome. 70 Para Para See text to n. 64 above. 13

14 instances where the protection of consumers interests requires that. The paper does not appear to recognise this but rather presumes that protecting the interests of the competitors would naturally result in the protection of consumers interests. With regard to the efficiency defence, the dominant undertaking must be able to demonstrate that the efficiencies brought about by the conduct outweigh the likely negative effects on competition resulting from it and hence the likely harm to consumers that the conduct might otherwise have. 73 This last statement resembles a rule of reason approach in that so long as the efficiencies brought about by the conduct outweigh the negative effects on competition and the possible harm to consumers, it may escape prohibition under Article 82EC. It is as such also similar to Article 81(3)EC in a narrower sense - narrower in that it is limited to the realisation of efficiency gains only. It is worth pointing out that it is not the competition authority or the court that weighs pro- versus anti-competitive effects of conduct, but the dominant company that is allowed to argue in that manner as a defence. The novelty is that the efficiency defence found in EC merger control 74 seems to be adopted in the context of abuse of a dominant position as well. The Commission thus appears to have followed the EAGCP Report which argues for assessment of the extent to which the competitive harm identified by the authority is potentially outweighed by efficiency gains. 75 Although the EAGCP Report is not very consistent as to whom the burden of proof falls 76, the paper explicitly states that it will be on the dominant company. 77 Nonetheless, the overall attitude of the paper towards efficiencies is more stringent than the EAGCP Report s in that the protection of rivalry and the competitive process is given priority over possible pro-competitive efficiency gains. 78 Furthermore, it is deemed highly unlikely that an abusive conduct of a dominant company with a market position approaching that of a monopoly or with a similar degree of market power could be justified on the ground that efficiency gains would 73 Para See Commission Guidelines on horizontal mergers [2004] OJ C31/5 paras EAGCP Report n. 2 above, p It is first stated in the EAGCP Report that competition authorities will need to identify a competitive harm, and assess the extent to which such a negative effect on consumers is potentially outweighed by efficiency gains (n. 2 above, p.3), later on it is argued that [c]ompetition authorities have to show the presence of siginificant anti-competitive harm, while the dominant firm should bear the burden of establishing credible efficiency arguments (n. 2 above, p. 4). 77 Para Para

15 counteract its actual or likely anti-competitive effects. 79 This is unfortunately another example of the loyalty to the current formalistic approach. For the efficiency defence, the dominant company has to demonstrate that efficiencies are realised or likely to be realised as a result of the conduct concerned; that the conduct is indispensable to realise these efficiencies; that the efficiencies benefit consumers and that competition in respect of a substantial part of the products concerned is not eliminated. 80 The reflection of Article 81(3)EC in these criteria almost suggests that an exemption clause under Article 82EC is to be brought by the Commission application of that provision. Specific Types of Exclusionary Abuse The paper examines some types of exclusionary behaviour in more detail. These are predatory pricing, single branding and rebates, tying and bundling and refusal to supply. Only a brief examination of these will be conducted here. Predatory Pricing As regards predatory pricing, it is made clear that it will only be dealt with as an abuse if the dominant company applies it to protect or strengthen its position (with the exception to this being the EC Commission s policy in sectors where activities are protected by legal monopoly). Predatory pricing in an unrelated market where the company is not dominant and where the predation will only have effects in this unrelated market will normally not be an abuse. 81 The paper notes that the dominant company will sacrifice by making short run losses when it considers that it is likely to be able to recoup the losses or lost profits at a later stage after foreclosure occurs. 82 Although at first sight, the paper can be understood to be bringing a condition of recoupment unlike the current state, it later on continues by asserting that the Commission does not consider it necessary to provide further separate proof of recoupment in order to find abuse. 83 Hence, the probability of recoupment is still not part of the predatory pricing test. The paper in this regard does not follow the EAGCP 79 Para Para Para The concern here is the prevention of cross subsidisation, ibid. The Commission cites the Tetra Pak II case (C-333/94 P Tetra Pak International SA v Commission [1996] ECR I-5951) at this point to clarify that the case was exceptional there in that the markets of aseptic and non-aseptic cartons were strongly linked and due its quasi monopolistic position on the first and leading position on the closely associated latter market, Tetra Pak had a quasi dominant position on the latter market as well. 82 Para Para The Commission seems to assume that once dominance is established, since this normally means that entry barriers are sufficiently high, the possibility to recoup may be presumed, ibid. 15

16 Report which argues that the recoupment ability of the predator has to be carefully assessed. 84 Another novelty that the paper introduces is a change in the appropriate cost benchmark for predatory pricing. Whereas the current benchmark adopted by the Commission and the courts is average variable cost 85, the paper establishes the concept of average avoidable cost and states the relevant question to be whether the dominant company, by charging a lower price for all or a particular part of its output over the relevant period of time incurred or incurs losses that could have been avoided by not producing that (particular part of its) output. 86 If the dominant company incurs such avoidable losses, then the pricing can be presumed to be predatory. The dominant company is allowed to rebut this presumption. Hence, it is made clear that pricing even below the relevant cost benchmark, i.e. now the average avoidable cost may be justified. 87 Such justification may be by way of showing that the dominant company is actually minimising its losses in the short run, for example, in case of restart up costs or strong learning effects. 88 However, a meeting competition defence can normally not be used where the dominant company is pricing below the average avoidable cost and an efficiency defence can in general not be applied to predatory pricing. 89 It is worthwhile to note that the new cost benchmark, i.e. the average avoidable cost would in certain instances be higher than the average variable cost 90 which implies that a finding of predation is made easier by the new approach. Moreover, this focus of the paper on the cost benchmark is not very much in line with the EAGCP Report which argues that the price-cost rules used in EC law do not necessarily provide convincing evidence of a sound predation story since the predator s costs are not that relevant for identifying the nature of concern. 91 Single Branding and Rebates 84 EAGCP Report n. 2 above, p See Case C-62/86 AKZO Chemie BV v Commission [1991] ECR I-3359; Tetra Pak II, n. 81 above. 86 Para See AKZO where the ECJ held that [p]rices below average variable costs by means of which a dominant undertaking seeks to eliminate a competitor must be regarded as abusive (n. 85 above, para. 71). 88 See paras for possible defences. 89 Paras For the argument that the dominant company should be able to sell below cost where its competitors are doing the same see Whish n. 23 above, p See para EAGCP Report n. 2 above, p

17 On single branding and rebates, the paper recognises that these systems can be used for both efficiency enhancing and anti-competitive reasons; thus, they may have efficiency enhancing effects as well as anti-competitive effects. 92 Possible efficiency explanations and defences are considered to be the case of relationship specific investments, costs advantages and avoidance of double marginalisation. 93 The possible negative effects are regarded to be foreclosure of the market and price discrimination between buyers. 94 In general, the rebate system should not hinder as efficient competitors to expand or enter. 95 A substantial part of the paper is dedicated to the assessment of rebates paying attention to their different forms and effects. 96 Tying and Bundling As for tying and bundling the possible anti-competitive effects are argued to be foreclosure, price discrimination and higher prices. 97 Moreover, a new concept of tying, i.e. technical tying is introduced. Accordingly, technical tying occurs when the tied product is physically integrated in the tying product. 98 This definition brings to mind the software industry and obviously the Microsoft decision where Microsoft was found to abuse its dominant position by tying Windows Media Player to the Windows Operating System. 99 The concept may, however, cause concern since when two products are physically integrated, proving that they are separate products becomes harder and the argument that integrating two products into one where it is possible is more efficient than having the two separately would be harder to rebut. According to the paper, the demand of the customers would determine whether two products are distinct or not; if, in the absence of tying or bundling, from the customers perspective, two products are or would be purchased separately, then they are distinct products. 100 One difficulty of this approach would be that when a new product is introduced with integrated features, it would be impossible to know whether there is or would have been demand for these features separately. Moreover, such an approach goes beyond Article 82(2)(d)EC since that provision requires a lack of connection between the tied products with regard to their nature or commercial 92 Para Paras. 172 et seq. 94 Paras Para See paras Para Para Commission Decision Microsoft 24 March 2004 COMP/C-3/ Para

18 usage for a finding of abuse. Hence, the approach seems to legitimise the holding in Tetra Pak II that even a natural link or commercial usage is not always enough for the tying to escape the prohibition in Article 82EC. 101 Refusal to Supply For refusal to supply to be abusive, the refusal must have a likely anti-competitive effect on the market which is detrimental to consumer welfare. 102 It is recognised that forcing companies to supply may affect investment incentives both negatively and positively. Hence, competition policy towards refusal to supply should take into consideration both the effect of having more short-run competition and the possible long-run investment incentives. 103 With regard to refusal to licence intellectual property rights, the paper first repeats the position taken by the ECJ in IMS Health 104 that a duty to licence can arise only if the undertaking requesting the licence does not intend to limit itself essentially to duplicating the goods or services already offered by the owner of the intellectual property right, but intends to produce new goods or services not currently offered and for which there is a potential consumer demand. 105 Nevertheless, the paper then suggests that a refusal to licence an intellectual property right protected technology indispensable for follow-on innovation by competitors may still be abusive even if the licence is not sought to directly incorporate the technology in clearly identifiable new goods and services. 106 This is a problematic assertion for two reasons. First, the ECJ judgment in IMS Health explicitly looks for the refusal to prevent the emergence of a new product for which there is a potential consumers demand 107 and does not differentiate between follow-on and other types of innovation. Thus, the Commission s approach in the paper seems to go beyond what the ECJ held and be broadening the scope of exceptional circumstances 108 under which a licence holder may be asked to licence 101 See Tetra Pak II, n. 81 above, para Para Para A duty to supply may lead companies not to invest or invest less to begin with and also result in the other companies free riding on the dominant company s investments. On the other hand, access to input may lead the other companies to increase their investment in follow-on research and development, ibid. 104 Case C-418/01 IMS Health GmbH & Co OHG v NDC Health GmbH & Co KG [2004] 4 CMLR 1543, para Para. 239 citing IMS Health, n. 104 above. 106 Para IMS Health, n. 104 above, para IMS Health, n. 104 above, para

19 its property. Such a position seems in conformity with its Microsoft decision 109, but in ignorance of the ECJ judgment in IMS Health held subsequently. Further to support its position in Microsoft, the Commission argues that in case of refusal to supply interoperability information, the usual high standards for intervention in refusal to licence intellectual property rights may not be appropriate; thus, intervention can be justified more easily. Secondly, such an approach may easily be exploited by competitors if they do not have to directly incorporate the technology in clearly identifiable new products and such loose language could result in the unduly reduction of innovation incentives. Aftermarkets The paper finally provides guidance on aftermarkets. 110 It is rightly recognised that if the primary market is competitive, competition there may make price increases in the aftermarket unprofitable due to its impact on sales in the primary market, unless prices in the primary market are lowered to offset the higher aftermarket price. 111 The paper notes that if an aftermarket consisting of the secondary products of one brand of primary product has been found to constitute the relevant product market, a dominant position on such a market can only be established after analysing competition on both the aftermarket and the primary market. 112 There is also a presumption of abuse that if a dominant position on an aftermarket has been established following the methodology in the paper, the Commission is to presume that it is abusive for the dominant company to reserve the aftermarket for itself by excluding competitors from the market. 113 The dominant company is nevertheless able to bring forward an objective justification such as the guarantee of the quality and good usage of the products or an efficiency defence. 114 Conclusion The paper is overall a welcome attempt in that it seems to be adopting a more effectsbased approach towards Article 82EC in general than the status quo and is proposing 109 There the Commission held that there was no persuasiveness to an approach that would favour an exhaustive checklist of exceptional circumstances, Microsoft, n. 99 above, para Paras Aftermarkets are those comprising complementary products ( secondary products ) that are purchased after the purchase of another product ( primary product ) to which it relates, para Para The market definition exercise is to focus on the aftermarket sales to customers who have already obtained the primary product, not on potential future buyers of the primary product, ibid Para Para Para